A dendroecological method to evaluate past caribou (Rangifer tarandus L.) activity

Records of population changes of caribou come from various sources (historical accounts, hunting and trading statistics, herd surveys) that are typically incomplete and discontinuous in time and space. Here we propose a new method for the evaluation of past caribou activity using tree-ring records from boreal and subarctic conifer stands. The age-frequency distributions of trampling scars produced by caribou hooves on surficial roots and low branches of erect and stunted conifers are used as an index of the passage of caribou through stands during the snow-free period. To verify if changes in the age structure of trampling scars correspond to changes in abundance of caribou movement, we analyzed factors influencing production and loss of scars at two lichen-woodland sites in northeastern Québec-Labrador (Canada). The detailed analysis of trampling scars in the first site indicates that the capacity of conifers to produce scars is maintained under a regime of repeated caribou traffic; scars were formed at new positions along the exposed roots and scars continued to be produced at a same position in a minimum time of one growing season, even after 15 years of caribou traffic. The influence of repeated caribou trampling on loss of scars was measured by comparing the age structure of scars of three vegetation groups (based on caribou trail network) with different intensities of use. The similarity of the age structures of the three groups showed that scar loss due to trampling was hardly detectable, which indicated that scar loss was low in comparison to the number of scars produced, even in the most used trails. Sampling of trampling scars during two successive years at the second site showed that the stability of the age structure of scars was not affected by moderate caribou traffic. Our results, therefore, indicate that most of the information deduced from the age structures of trampling scars comes from changes in caribou activity. The method opens the possibility of assessing caribou activity in time (several decades) and space over large areas of the boreal forest and the forest-tundra biomes. Notre connaissance de la dynamique des populations de caribou provient de diverses sources (documents historiques, données sur la chasse et le commerce, inventaire des troupeaux) qui livrent une information incomplète et discontinue dans l'espace et dans le temps. Nous présentons ici une méthode dendroécologique qui permet d'évaluer la fréquentation passée des mil